Floer homology for magnetic fields with at most linear growth on the universal cover (Q408138)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Floer homology for magnetic fields with at most linear growth on the universal cover |
scientific article |
Statements
Floer homology for magnetic fields with at most linear growth on the universal cover (English)
0 references
29 March 2012
0 references
One important result of the last decade in symplectic topology is the isomorphism between Floer homology of a cotangent bundle and the singular homology of the free loop space of the underlying closed oriented manifold. It was independently proved by \textit{A. Abbondandolo} and \textit{M. Schwarz} [Commun. Pure Applied Math. 59, No. 2, 254--316 (2006; Zbl 1084.53074)] and by \textit{D. A. Salamon} and \textit{J. Weber} [Geom. Funct. Anal. 16, No. 5, 1050--1138 (2006; Zbl 1118.53056)]. The goal of the paper under review is a generalization of this theorem for the case of twisted cotangent bundles, i.e. \(T^{\ast}M\) endowed with the two-form \(\omega_{\sigma}=\sum_{i}dp_{i}\wedge dq_{i}\, +\, \pi^{\ast}\sigma\), where \(\sigma\) is a closed two-form on \(M\) and \(\pi:T^{\ast}M\rightarrow M\) is the projection. Some hypotheses are needed in order to have a well-defined Floer homology in this framework: {\parindent=6mm\begin{itemize}\item[(i)] The lift \(\tilde{\sigma}=p^{\ast}\sigma\) on the universal cover \(p:\widetilde{M}\rightarrow M\) is exact. This is needed to prevent the occurence of bubbling holomorphic spheres.\item[(ii)] There is a primitive \(\theta\) of \(\tilde{\sigma}\) with at most linear growth, i.e. for some Riemanian metric on \(M\) the norm of \(\theta\) measured with the lifted metric grows at most linearly with respect to the radius of the geodesic ball centered at some point of \(\widetilde{M}\). This condition is used to establish \(L^{\infty}\)-bounds on the perturbed holomorphic curves in order to prove compactness. \end{itemize}} Under these assumptions the authors define the Floer homology groups \(HF^{\alpha}_{\ast}(H,\delta\sigma, \tau)\). Here, \(H\) is a \(\tau\)-periodic Hamiltonian satisfying the same growth conditions as in the paper of Abbondandolo-Schwartz, \(\delta\in (0, \delta_{0}/\tau)\) for some constant \(\delta_{0}\), and \(\alpha\) is a free homotopy class in \(M\) which is \(\sigma\)-atoroidal, meaning that \(\int_{T}i^{\ast}\sigma=0\) for any torus in \(i:T=S^{1}\times S^{1}\rightarrow M\) such that \(i\) maps the two factors \(S^{1}\) in loops belonging to \(\alpha\). They prove that this Floer homology is isomorphic to the singular homology of the space of \(\tau\)-periodic loops belonging to \(\alpha\). The authors also point out that \(\delta_{0}\) equals \(+\infty\) when there is a \textit{bounded} primitive \(\theta\) of \(\tilde{\sigma}\): According to \textit{M. Gromov} [J. Differ. Geom. 33, No. 1, 263--292 (1991; Zbl 0719.53042)], this is the case when \(M\) admits a metric of negative curvature. Under this hypothesis they prove the Conley conjecture for the corresponding twisted cotangent bundle: A time one map of a Hamiltonian satisfying the Abbondandolo-Schwarz growth conditions which has finitely many fixed points, has simple periodic orbits of arbitrarily large period.
0 references
Floer homology
0 references
closed orbits
0 references
Conley conjecture
0 references
\(L^{\infty }\) estimates
0 references
twisted cotangent bundle
0 references
Palais-Smale condition
0 references
0 references
0 references
0 references
0 references